Taku Ito
LightNews LightNews
banner
takuito.bsky.social
Taku Ito
@takuito.bsky.social
590 followers 540 following 5 posts

Computational Neuroscience + AI @ IBM Research | 📍NYC | https://ito-takuya.github.io

Videos
Neuroscience 41%
Engineering 35%
Posts Replies Media Videos
Pinned
takuito.bsky.social
Taku Ito @takuito.bsky.social · Aug 19
What complexity of algorithms can AI compute? In a new paper with colleagues at IBM Research, we explore how circuit complexity theory can help quantify the degree of algorithmic generalization in AI systems. www.nature.com/articles/s42...
@natmachintell.nature.com
#ML #AI #MLSky
1/n

Reposted by Stephen M. Walt, Joanna Bryson, Marc Lynch , and 80 more

davidho.bsky.social
David Ho @davidho.bsky.social · 20d
This is the most astonishing graph of what the Trump regime has done to US science. They have destroyed the federal science workforce across the board. The negative impacts on Americans will be felt for generations, and the US might never be the same again.

www.nature.com/immersive/d4...
This line graph illustrates the percentage change in agency staff levels from the previous year for nine major U.S. federal scientific and health organizations between the fiscal years 2016 and 2025. The agencies tracked include the CDC, Department of Energy, EPA, FDA, NASA, NIH, NIST, NOAA, and NSF. For the majority of the timeline between 2016 and 2023, the agencies show relatively stable fluctuations, generally staying within a range of +5% to -5% change per year. However, there is a dramatic and uniform plummet starting in the 2024–25 period. Every agency depicted shows a sharp downward trajectory, with staffing losses ranging from approximately -15% to over -25%. The Environmental Protection Agency (EPA) shows the most significant decline, dropping to roughly -26%, while the National Institute of Standards and Technology (NIST) shows the least severe but still substantial drop at approximately -15%.

Reposted by Brian A. Nosek, Joshua S. Weitz, Juan Rocha , and 2 more

maxkozlov.bsky.social
Max Kozlov @maxkozlov.bsky.social · 21d
Trump has been in office for one year. We at @nature.com did a deep dive looking at the administration's disruption of science in numbers.

Take a look—the numbers are staggering. By me, @dangaristo.bsky.social, Jeff Tollefson, @kimay.bsky.social, & help from @noamross.net @scott-delaney.bsky.social
US science after a year of Trump: what has been lost and what remains
A series of graphics reveals how the Trump administration has sought historic cuts to science and the research workforce.
www.nature.com

Reposted by Taku Ito

eugenevinitsky.bsky.social
Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 29d
Oh wow, deepseek is starting to make serious progress on LLMs that offload memory to external storage: github.com/deepseek-ai/...
github.com

Reposted by Joanna Bryson, Taku Ito, Gilles Louppe

sakanaai.bsky.social
sakanaai.bsky.social @sakanaai.bsky.social · 29d
Introducing DroPE: Extending Context by Dropping Positional Embeddings

We found embeddings like RoPE aid training but bottleneck long-sequence generalization. Our solution’s simple: treat them as a temporary training scaffold, not a permanent necessity.

arxiv.org/abs/2512.12167
pub.sakana.ai/DroPE

Reposted by Taku Ito

ravimill.bsky.social
Ravi Mill @ravimill.bsky.social · Nov 19
Excited to see our paper with @mwcole.bsky.social finally out in peer-reviewed form @natcomms.nature.com! We examine how the human brain learns new tasks and optimizes representations over practice…1/n
Schematic depicting cortical-subcortical interactions during multi-task learning

Reposted by Taku Ito

engeltatiana.bsky.social
Tatiana Engel @engeltatiana.bsky.social · Oct 24
Our work with @pawa-pawa.bsky.social is out in Nature Machine Intelligence! The choice of activation function affects the representations, dynamics, and circuit solutions that emerge in RNNs trained on cognitive tasks. Activation matters!
www.nature.com/articles/s42...

Reposted by Taku Ito

meharpist.bsky.social
Mary Elizabeth Sutherland @meharpist.bsky.social · Oct 24
Did you know that AI can figure out its own way to learn, and that its way is better than one designed by humans? Read more in a @nature.com N&V (and the original paper is in the comment) 🧪 www.nature.com/articles/d41...
AI discovers learning algorithm that outperforms those designed by humans
An artificial-intelligence algorithm that discovers its own way to learn achieves state-of-the-art performance, including on some tasks it had never encountered before.
www.nature.com

Reposted by Larry W. Hunter, Taku Ito

ramon-astudillo.bsky.social
Ramon Astudillo @ramon-astudillo.bsky.social · Oct 7
(repost welcome) The Generative Model Alignment team at IBM Research is looking for next summer interns! Two candidates for two topics

🍰Reinforcement Learning environments for LLMs

🐎Speculative and non-auto regressive generation for LLMs

interested/curious? DM or email ramon.astudillo@ibm.com

Reposted by Taku Ito

pessoabrain.bsky.social
PessoaBrain @pessoabrain.bsky.social · Oct 6
Michael X Cohen on why he left academia/neuroscience.
mikexcohen.substack.com/p/why-i-left...
Why I left academia and neuroscience
Don't worry, this isn't yet another story of rage-quitting.
mikexcohen.substack.com

Reposted by Taku Ito

nature.com
Nature @nature.com · Sep 26
Nature research paper: Arousal as a universal embedding for spatiotemporal brain dynamics

go.nature.com/4nMUgYz
Arousal as a universal embedding for spatiotemporal brain dynamics - Nature
Reframing of arousal as a latent dynamical system can reconstruct multidimensional measurements of large-scale spatiotemporal brain dynamics on the timescale of seconds in mice.
go.nature.com

Reposted by Alessandro Gozzi, Taku Ito

mwcole.bsky.social
Michael W. Cole @mwcole.bsky.social · Sep 14
Lab’s latest is out in Imaging Neuroscience, led by Kirsten Peterson: “Regularized partial correlation provides reliable functional connectivity estimates while correcting for widespread confounding”, where we demonstrate a major improvement to standard fMRI functional connectivity (correlation) 1/n

takuito.bsky.social
Taku Ito @takuito.bsky.social · Aug 19
Formalizing AI computation in terms of algorithmic complexity can offer a formal way to quantify AI systems while offering a principled foundation to build more algorithmically capable systems in the future.
Blog: research.ibm.com/blog/ai-algo...
arXiv: arxiv.org/abs/2411.05943
Can AI generate truly novel algorithms?
A decades-old approach to measuring algorithmic complexity could provide a window into better understanding how AI systems compute.
research.ibm.com

takuito.bsky.social
Taku Ito @takuito.bsky.social · Aug 19
While using AI models to generate code is commonplace these days, we still do not fully understand the limits of the complexity of the code these models can formulate.
3/n

takuito.bsky.social
Taku Ito @takuito.bsky.social · Aug 19
Using circuits to formalize algorithmic problems for AI models (e.g., depth as time complexity, size as space complexity), we can quantify the complexity of circuit computations (algorithmic complexity) an AI model can perform.
2/n

Reposted by Taku Ito

milhammichael.bsky.social
milhammichael.bsky.social @milhammichael.bsky.social · Mar 7
Mental health research is at a turning point—breakthroughs can transform lives, but only with bold action, investment, and open collaboration. The time for action is now. Read our full statement here: childmind.org/blog/can-sci...

Reposted by Todd M. Gureckis, Taku Ito

guydav.bsky.social
Guy Davidson ✈️ NeurIPS 2025 @guydav.bsky.social · Feb 21
Out today in Nature Machine Intelligence!

From childhood on, people can create novel, playful, and creative goals. Models have yet to capture this ability. We propose a new way to represent goals and report a model that can generate human-like goals in a playful setting... 1/N

Reposted by Taku Ito

nhiratani.bsky.social
Naoki Hiratani @nhiratani.bsky.social · Feb 6
New preprint! Ziyan and I explore how task order impacts continual learning in neural networks and how to optimize it. Our analysis highlights two key principles for better task sequencing.
Check it out: arxiv.org/pdf/2502.03350
arxiv.org

Reposted by Sarah Mitchell, Taku Ito

docbecca.bsky.social
Dr. Becca @docbecca.bsky.social · Jan 31
The entire website for the NIH Office of Research on Women's Health (ORWH) is very nearly stripped bare. This is so, so devastating. orwh.od.nih.gov/research/fun...
orwh.od.nih.gov

Reposted by Taku Ito

veithweilnhammer.bsky.social
Veith Weilnhammer, MD @veithweilnhammer.bsky.social · Jan 19
New paper in @brain1878.bsky.social: Healthy people under S-ketamine, an NMDAR antagonist, and people living with schizophrenia, a disorder associated with NMDAR hypofunction, spend more time in an external mode of perception - where noisy sensory signals override knowledge about the world.

Reposted by Taku Ito

jsaito25.bsky.social
Joey Saito @jsaito25.bsky.social · Jan 16
The origin of color categories | PNAS www.pnas.org/doi/10.1073/...
The origin of color categories | PNAS
To what extent does concept formation require language? Here, we exploit color to address this question and ask whether macaque monkeys have color ...
www.pnas.org

Reposted by Taku Ito

biorxiv-neursci.bsky.social
bioRxiv Neuroscience @biorxiv-neursci.bsky.social · Jan 11
Quantifying Differences in Neural Population Activity With Shape Metrics https://www.biorxiv.org/content/10.1101/2025.01.10.632411v1

Reposted by Taku Ito

emollick.bsky.social
Ethan Mollick @emollick.bsky.social · Jan 11
Paper shows very small LLMs can match or beat larger ones through 'deep thinking' - evaluating different solution paths - and other tricks. Their 7B model beats o1-preview on complex math by exploring 64 different solutions & picking the best one.

Test-time compute paradigm seems really fruitful.

Reposted by Taku Ito

juliencorbo.bsky.social
Julien Corbo @juliencorbo.bsky.social · Jan 8
New paper out! 🚨 📰 With @batuhanerkat.bsky.social, John McClure, @hussainyk1.bsky.social, @polacklab.bsky.social we reveal how discretized representations in V1 predict suboptimal orientation discrimination. 🧪🧠🐭 This work reconciles neuro and psychometric curves
www.nature.com/articles/s41...
Discretized representations in V1 predict suboptimal orientation discrimination - Nature Communications
How animals generate perceptual decisions remains poorly understood. Here, the authors show that during a discrimination task, the mouse visual cortex does not encode the orientations of the cues but ...
www.nature.com

Reposted by Taku Ito

marlenecohen.bsky.social
Marlene Cohen @marlenecohen.bsky.social · Jan 4
New results for a new year! “Linking neural population formatting to function” describes our modern take on an old question: how can we understand the contribution of a brain area to behavior?
www.biorxiv.org/content/10.1...
🧠👩🏻‍🔬🧪🧵
#neuroskyence
1/
Linking neural population formatting to function
Animals capable of complex behaviors tend to have more distinct brain areas than simpler organisms, and artificial networks that perform many tasks tend to self-organize into modules (1-3). This sugge...
www.biorxiv.org

Reposted by Taku Ito

tyrellturing.bsky.social
Blake Richards @tyrellturing.bsky.social · Jan 3
And relatedly, Felix wrote a good piece on the stress and anxiety currently affecting many people who work in AI due to the current climate in the industry:

docs.google.com/document/d/1...

If only more folks in AI were gentle and introspective like this...
AI and Stress
200Bn Weights of Responsibility The Stress of Working in Modern AI Felix Hill, Oct 2024 The field of AI has changed irrevocably in the last 2 years. ChatGPT is approaching 200m monthly users. Gemin...
docs.google.com

Reposted by Taku Ito

David Bau @davidbau.bsky.social · Dec 31
What was the most important machine learning paper in 2024?

My Famous Deep Learning Papers list (that I use in teaching) does not include any new ideas from the last year.

papers.baulab.info

Which single new paper would you add?

Reposted by Arvind Narayanan, Sendhil Mullainathan, Arun Sundararajan , and 1 more

melaniemitchell.bsky.social
Melanie Mitchell @melaniemitchell.bsky.social · Dec 23
Some of my thoughts on OpenAI's o3 and the ARC-AGI benchmark

aiguide.substack.com/p/did-openai...
Did OpenAI Just Solve Abstract Reasoning?
OpenAI’s o3 model aces the "Abstraction and Reasoning Corpus" — but what does it mean?
aiguide.substack.com

Reposted by Taku Ito

xiaoxuanlei.bsky.social
Xiaoxuan @xiaoxuanlei.bsky.social · Dec 12
📌 Poster Session:
⏰ When: TODAY, Thu, Dec 12, 4:30 p.m. – 7:30 p.m. PST
📍 Where: East Exhibit Hall A-C, #3705
📄 What: Geometry of Naturalistic Object Representations in Recurrent Neural Network Models of Working Memory

Hope to see you there!
@bashivan.bsky.social @takuito.bsky.social

Reposted by Stanislas Dehaene, Francisco C. Pereira, Taku Ito

jeanremiking.bsky.social
Jean-Rémi King @jeanremiking.bsky.social · Dec 12
🚨We're very excited to share our latest study, by Pablo Diego and team:

"A polar coordinate system represents syntax in large language models",

📄: Paper arxiv.org/abs/2412.05571
🪧: Poster tomorrow: neurips.cc/virtual/2024...
🧵: Thread 👇